10DBMC International Conférence On Durability of Building Materials and Components

LYON [France] 17-20 April 2005

Concrete Deterioration Caused by Sulfuric Acid Attack

K. Kawai, S. Yamaji, T. Shinmi

Department of Social and Environmental Engineering,
Graduate School of Engineering, Hiroshima University
1-4-1, Kagamiyama, Higashi-Hiroshima, 739-8527 Japan
kkawai@hiroshima-u.ac.jp

TT1-199

ABSTRACT

Recently, biological deterioration of concrete in sewage and wastewater treatment plants has been
reported. This deterioration is caused by sulfuric acid attack and is dependent on the concentration of
sulfuric acid, this in turn being a function of both the specific location within the plant and also the
time over which the concrete is exposed to elevated concentrations of acid. Given that concrete may
often be exposed to very strong acid solutions, resin coatings are applied to the concrete to protect
them. However these coating are only effective if there are neither pinholes nor defects due to coating
operations. Otherwise, the deterioration of the concrete may proceed rapidly by sulfuric acid and
sulfate. The method of predicting deterioration of coated concrete when subjected to sulfuric acid and
sulfate attack has not yet been established. It is understood that the depth of deterioration of concrete
due to sulfuric acid attack is proportional to the square root of the exposure time. This relationship is
based on results of laboratory tests where specimens are continuously immersed in acid solutions over
a specfied period of time. But in actual structures, concrete is often exposed to flowing water that
contains an acid solution such as sewage. In other words, concrete is subjected to the shearing force of
fluid that erodes the surface of areas of deteriorated concrete. In these instances, it is predicted that
concrete deterioration caused by sulfuric acid attack will proceed at a greater rate than that predicted
from the square root relationship. Hence in this study concrete deterioration caused by sulfuric acid
attack was investigated considering the effects of the flow of acid solution over the surface of concrete
with the intent of proposing a prediction method for the deterioration of concrete due to sulfuric acid.
Cylindrical concrete specimens and mortar prisms were immersed in various concentrations of sulfuric
acid. In certain tests the sulfuric acid solution was circulated onto concrete specimens. In both
instances, the depths of zones eroded and neutralized by acids were measured. As well, the zones of
deteriorated concrete were analysed with an XRD and an ion chromatoanalyzer. It was found that the
rate of concrete deterioration caused by sulfuric acid attack depended on the pH value of acid solutions
and that the depth of erosion of concrete was nearly proportional to the exposure time of flowing acid
solution to which concrete was exposed.

KEYWORDS

Concrete, Chemical Deterioration, Acid Attack, Sulfuric Acid, Flowing Water

 10DBMC International Conférence on Durability of Building Materials and Components
LYON [France] 17-20 April 2005

1 INTRODUCTION

Sulfuric acid solution in sewage, wastewater treatment plants and hot spring places deteriorates
concrete structures hard by reacting with cement hydrates. Concrete is not a chemically stable material
under the condition of acidic environment. Although acid attack is one of the primary chemical
deterioration of concrete for many years, the prediction method for this kind of deterioration has not
been established yet. In Japan, the JSCE (Japan Society of Civil Engineers) Standard Specification
[2001] mentions about countermeasures against chemical deterioration and mentenances, but does not
descirbe the prediction method.

As a study of the prediction method for deterioration by sulfuric acid attack, many kinds of immersion
tests using mortar specimens have been carried out. But actual structures are made from concrete and
concrete deterioration due to acid attack should be predicted. Those studies are not so much so far.

It is generally said that the deterioration depth of concrete due to acid attack including sulfuric acid
attack is proportional to the square root of the exposure time, according to Sakamoto [1972], Pavlik
[1994], and so on, as follows:

y=b t

where, y the deterioration depth (mm)

t the exposure time in acid solution (year)
b constant (mm/ year ).

In this equation, the rate of deterioration is governed by the diffusion rate of acid in deteriorated zones
from concrete surfaces to non-deteriorated concrete under the assumption that corrosion products keep
to remain on concrete surfaces. In actual deteriorated structures of sewage and wastewater treatment
plants, however, it is found that deterioration products have been removed from concrete surfaces and
coarse aggregates are exposed on surfaces. This should be because the deterioration products of
mainly gypsum are subjected to the shearing force of flowing water and/or splashing water.

Usually experimental studies on concrete deterioration due to sulfuric acid attack are performed using
specimens statically immersed in acid solutions. In this case, because concrete surfaces are not
subjected to the shearing force, deterioration products will remain on the surfaces. It is concerned that
the results obtained from these experiments could evaluate lesser deterioration than actual
deterioration of sewage and wastewater treatment plants subjected to the shearing force of folwing
water.

In this study, in order to clarify the effects of the shearing force of flowing water on the concrete
deterioration due to sulfuric acid attack, concrete and mortar specimens are immersed in sulfuric acid
solutions that are circulated by pumps. This circulation generates the flow of solutions and the
shearing force of the flow is applied to the surfaces of the specimens.

2 EXPERIMENTAL PROCEDURES

2.1 Materials and specimens

Ordinary Portland cement was used as cement. Blast furnace slag and fly ash were used as mineral
admixtures. Concrete cylinders of φ150×300 mm in size and mortar prisms of 40×40×160 mm in size
were made with water binder ratios of 0.35, 0.50 and 0.65. The replacement ratio of cement with
mineral admixtures was 0.30. The propoerties of materials used in this study and the mix proportions
are shown in Table 1 and Table 2, respectively. Comparing the chemical resistivity at a certain age, it

TT1-199, Concrete deterioration caused by sulfuric acid attack by K. Kawai, S. Yamaji and T. Shinmi
 10DBMC International Conférence on Durability of Building Materials and Components
LYON [France] 17-20 April 2005

Table 1. Physical properties of materials used in this study

Ordinary Portland cement
Cement
Density : 3.16 (g/cm3), Blaine fineness : 3080 (cm2/g)
Fly ash
Mineral Density : 2.30 (g/cm3), Blaine fineness : 4160 (cm2/g)
admixtures Blast furnace slag
Density : 2.91 (g/cm3), Blaine fineness : 6220 (cm2/g)
Iwase crushed stone
Coarse
Maximum Size : 20 (mm), Density (SSD) : 2.65 (g/cm3)
aggregate
F.M. : 6.52, Water absorption : 0.74%, Solid content : 61.2%
Fine Ogasa crushed sand
aggregate Density (SSD) : 2.61 (g/cm3), F.M. : 2.82, Water absorption : 1.44%

Table 2. Mix proportions and compressive strengths of specimens used in this study
Percentage of Compressive
Replacement (%) Strength
Mark Binder W/
(Replaced by [N/mm2]
Mass) (Age: days)
NC35 ― 0.35 65.5 ( 28)
Ordinary Portland
NC50 ― 0.50 43.5 ( 28)
cement
NC65 ― 0.65 32.6 ( 28)
Concrete Ordinary Portland
BS30-65 cement + Blast furnace 30 0.65 36.3 ( 42)
slag
FA30-50 Ordinary Portland 30 0.50 43.1 ( 91)
FA30-65 cement + Fly ash 30 0.65 33.7 (417)
NC35 Ordinary Portland ― 0.35 47.2 ( 28)
Mortar
NC50 cement ― 0.50 41.3 ( 28)

is supposed that specimens containing mineral admixtures should be weaker than those containing no
mineral admixtures because of slow hydration. In this study, therefore, specimens containing mineral
admixtures were cured until those compressive strengths became almost equal to specimens
containing no mineral admixtures with the same water cement ratio cured for 28 days. It is assumed
here that the pore structures in specimens are almost same if compressive strengths are same with a
certain water cement ratio whether mineral admixtures are contained or not. The test results of
compressive strength of specimens are also shown in Table 2.

2.2 Experimental method

The concentrations of sulfuric acid solutions in immersion tests were 2.0 mol/L for concrete specimens
and about 0.09 mol/L (adjusted to pH=1.0) and about 0.06 mol/L (adjusted to pH=2.0) for mortar
specimens. The immersion tests contain two types of Method 1 and Method 2. In the Method 1,
specimens were immersed statically in sulfuric acid solutions and in the Method 2 they were immersed
in sulfuric acid solutions that were circulated with a pump and flowed in a tank (Table 3). To keep the
concentration of sulfuric acid solution constant, sulfuric acid was added suitably to the solution. At
this time, a pH meter was used to measure the value of pH and the concentration of sulfuric acid
solution was adjusted to the initial value of pH.

After the immersion tests were started, the erosion depth was measured with a vernier micrometer
every seven days. The erosion depth is defined as a distance between the initial surface and current

TT1-199, Concrete deterioration caused by sulfuric acid attack by K. Kawai, S. Yamaji and T. Shinmi
 10DBMC International Conférence on Durability of Building Materials and Components
LYON [France] 17-20 April 2005

Table 3. Methods for immersing specimens in sulfuric acid solutions

Abbreviation Detail
Method 1 Specimens are immersed statically in sulfuric acid solution in a tank.
Specimens are immersed in a tank where sulfuric acid solution is circulated
Method 2 by a pump. (Specimens are subjected to the shearing force of the flow of the
solution.)

surface. Before every measurement, intentional removal of deteriorated zones on the surfaces was not
carried out.

After specimens were immersed for certain periods, mineralogical changes in the specimens were
analyzed with an XRD and the distribution of sulfate ions in the specimens was measured with an ion
chromatoanalyzer.

3 RESULTS AND DISCUSSIONS

3.1 Effects of the flow of fluid on erosion

NC35 2.0mol/l
The erosion depth of concrete specimens NC50 black:M ethod 1
Erosion depth mm

immersed in 2.0 mol/L of sulfuric acid NC65 white:M ethod 2

solution with the Method 1 and Method 2 is 20 NC35
NC50
shown in Fig. 1. For specimens with every NC65
water cement ratio, the erosion depth with
the Method 2 was larger than with the 10
Method 1. Especially it is remarkable for
specimens of NC50 and NC65 after 60 day
immersion. As shown in Fig. 2, reaction 0
products of gypsum remained on the surfaces
of concrete specimens immersed with the 0 20 40 60 80 100
Method 1, while any products could not be Duration (days)
During(days)
seen on the surfaces of concrete specimens
immersed with the Method 2. The reaction Figure 1. Erosion depth of concrete immersed
in 2.0 mol/L of sulfuric acid solution
mm j

NC35 pH 1.0
NC50 black:Method 1
NC35 white:Method 2
Erosion depth i

20 NC50

10

0
20mm 0 20 40 60 80 100
During(days)
Duration (days)
Figure 2. Reaction products on the surface of
concrete specimen NC50 immersed Figure 3. Erosion depth of mortar immersed in
in 2.0 mol/L of sulfuric acid solution sulfuric acid solution of pH=1.0

TT1-199, Concrete deterioration caused by sulfuric acid attack by K. Kawai, S. Yamaji and T. Shinmi
 10DBMC International Conférence on Durability of Building Materials and Components
LYON [France] 17-20 April 2005

products on the surfaces of concrete specimens immersed with the Method 2 should be driven out by
the flow of fluid without being precipitated on the surfaces. As a result, sulfuric acid always attacks
concrete surfaces without the necessity of diffusing into deteriorated zones. Reaction products could
not be seen on the surfaces of concrete specimens NC35 immersed with the Method 1. Because of
dense microstructures, it is thought that reaction products of gypsum could not be precipitated in
microstructures and was driven out by the flow of fluid.

The erosion depth of mortar specimens immersed in ca. 0.09 mol/L (pH=1.0) of sulfuric acid solution
with the Method 1 and Method 2 is shown in Fig. 3. For NC35, erosion of specimens can be seen after
49 day immersion with the Method 1 and after 28 day immersion with the Method 2. For NC50,
erosion of specimens immersed with the Method 2 can be seen after 63 day immersion. The XRD
patterns of surface areas for mortar specimens of NC50 immersed for 63 days are shown in Fig. 4.
Gypsum was clearly identified for specimens immersed with both the Method 1 and Method 2. These
results show that reaction products of sulfuric acid attack are mainly gypsum. On the other hand, as
shown in Fig. 5, for mortar specimens of NC35 immersed in ca. 0.09 mol/L (pH=1.0) of sulfuric acid
solution for 63 days no reaction products can be seen on the surface of the mortar specimen immersed
with Method 2, while a layer of reaction products can be seen on the surface of the mortar specimen
immersed with the Method 1. Since reaction products were precipitated on the surface when the
specimen was immersed statically, reaction products would have been seen on the surface also in the
case of Method 2. But the flow of the solution swept reaction products on the surface.

1000 The erosion depth of mortar specimens immersed

Method 1 G GFypsum
G P FP ortlandite in ca. 0.06 mol/L (pH=2.0) of sulfuric acid
800 Q
solution with the Method 1 and Method 2 is shown
Q QFuartz
P P in Fig. 6. Every specimen immersed in sulfuric
600
G acid solution of pH=2.0 was hardly eroded but
400 swelled a little. Reaction products were found on
X-ray intensity (cps)

200
mm j

pH 2.0
NC35 black:Method
NC50 1
0 NC35 white:Method 2
G Gypsum
F
Erosion depth i

Method 2 20 NC50
P F
P ortlandite
800 Q Q Quartz
F
G P G P
600 10
400

200 0

0 0 20 40 60 80 100
10 20 30 40 50
2ƒ Æ
(deg) During(days)
Duration (days)
Figure 4. XRD patterns of surface areas Figure 6. Erosion depth of mortars immersed
for mortar specimens at the age in sulfuric acid solution of pH=2.0
of 28days (NC50)

Method 1 Method 2

NC35
Method 1 Method 2
NC50
Figure 5. Deterioration of NC35 immersed in Figure 7. Deterioration of NC35 and NC50
sulfuric acid solution of pH=1.0 for immersed in sulfuric acid solution
63 days of pH=2.0 for 63 days

TT1-199, Concrete deterioration caused by sulfuric acid attack by K. Kawai, S. Yamaji and T. Shinmi
 10DBMC International Conférence on Durability of Building Materials and Components
LYON [France] 17-20 April 2005

every surface of the specimens as shown in Fig. 7. But compared with specimens immersed in ca. 0.09
mol/L (pH=1.0) of sulfuric acid solution shown in Fig. 5, reaction products were not remarkable. The
specimens immersed in ca. 0.06 mol/L (pH=2.0) of sulfuric acid solution were not reacted heavier
than those immersed in ca. 0.09 mol/L (pH=1.0) of sulfuric acid solution.

Regarding the mechanism of concrete deterioration caused by sulfuric acid, Kurashige [2002]
descirbed that sulfuric acid penetrating into mortar or concrete reacts with calcium hydroxide of
cement hydrates to produce gypsum and at this time the volume of solid substances increases largely,
which causes expansion of reaction products resulting in erosion. This phenomenon is schematically
illustrated in Fig. 8. Concrete with a high water cement ratio has larger and more pores than that with
a low water cement ratio. These pores play the role of a capacity to absorb expansion caused by the
production of gypsum. Therefore concrete with a high water cement ratio has a higher capacity to
absorb the expansion of production reaction of gypsum than that with a low water cement ratio, that is
to say, concrete with a low water cement ratio erodes earlier than that with a high water cement ratio
and its erosion depth is nearly proportional to the exposure time instead of the square root of the
exposure time. The results of this study also agreed with this mechanism. It is shown that specimens
with a lower water cement ratio eroded heavier regardless of the Method 1 and Method 2 and the
erosion depth was nearly proportional to the exposure time.

Ueda et al. [1996] pointed out that sulfuric acid is hard to penetrate into hardened cement. As shown
in Fig. 9, sulfate ions did not penetrate very much into mortar or concrete also in this study. The
reaction of cement hydrates and sulfuric acid should occur only in the surface portion of specimens.
This would be because the reaction of production of gypsum in the surface portion is faster than the
penetration of sulfate ions into the specimen. The surface portion, therefore, is a main field of the
reaction of sulfuric acid. That is why specimens immersed with the Method 2 eroded much larger than
those with the Method 1 since the flow of solution removed the reaction product of gypsum.

3.2 Effects of the addition of mineral admixtures on erosion

The erosion depth of concrete specimens with mineral

admixtures immersed in 2.0 mol/L of sulfuric acid
solution with the Method 1 is shown in Fig. 10. The 30
Mortar 28days
Sulfate ion content in cement pastes [%]

erosion depth of specimens containing blast furnace Method 2 63days

slag and fly ash was smaller than that of specimens 20
containing no mineral admixtures. This is because the 28days
Erosion surface
production content of calcium hydroxide in concrete 63days
10 Erosion surface

High W/C Low W/C

0
Concrete 28days
Method 2 91days
20
28days
Erosion surface

10 91days
Erosion surface

0
0 10 20 30 40
Micro pores Distance from initial surface[ ‡ o
]
Hydrates
Figure 9. Sulfate ion content in cement
Figure 8. Mechanism of concrete deterioration paste portion of concrete and
due to sulfuric acid attack mortar (NC35)

TT1-199, Concrete deterioration caused by sulfuric acid attack by K. Kawai, S. Yamaji and T. Shinmi
 10DBMC International Conférence on Durability of Building Materials and Components
LYON [France] 17-20 April 2005

containing mineral admixtures was lower

mm j
than that in concrete containing no mineral 2.0mol/l
NC50 Method 1
admixtures since a part of cement was NC65
replaced with the mineral admixtures. As BS30-65

Erosion depth i
for concrete containing fly ash, since 20 FA30-50
FA30-65
calcium hydroxide produced by cement
hydration was consumed by the pozzolanic
reaction, the erosion depth was smallest 10
among concrete specimens with the same
water cement ratio. Consequently the
calcium hydroxide content in concrete is 0
closely related to the volume increase due
to the production of gypsum which causes 0 20 40 60 80 100
the deterioration and is one of the factors During(days)
Duration (days)
governing the degree of concrete
deterioration caused by sulfuric acid. Figure 10. Erosion depth of concrete containing
blast furnace slag and fly ash immersed
in 2.0 mol/L of sulfuric acid solution
4 CONCLUSIONS

This study was perfomed to understand the mechanism of concrete deterioration caused by sulfuric
acid. The effects of the flow of fluid, the concentration of sulfuric acid solution, the use of mineral
admixtures and the difference of water cement ratio on the deterioration of concrete were investigated.
As a result, the following conclusions were obtained.

1. Regarding concrete deterioration caused by sulfuric acid, the flow of fuid accelerates the
deterioration and the rate of deterioration of concrete caused by sulfuric acid strongly depnds upon
the concentration of sulfuric acid solution.. In an elevated concentration of sulfuric acid solution, the
erosion depth of concrete is nearly proportional to the exposure time instead of the square root of the
exposure time. Sulfate ions does not penetrate into concrete very much and the reaction of cement
hydrates and sulfuric acid occurrs in the surface portion of concrete.

2. When a part of cement is replaced with blast furnace slag or fly ash and the strength of concrete
containing the mineral admixture is almost equal to the strength of plain concrete with the same
water binder ratio, the erosion depth of concrete containing mineral admixture due to sulfuric acid
attack is smaller than that of plain concrete since the content of calcium hydroxide is small.

5. REFERENCES

JSCE, 2001, JSCE Standard Specifications for Concrete Structures [Mentenance], Japan Society of
Civil Engineers, Japan. (in Japanese)
Sakamoto, H. 1972, ‘Studies on resistance of cement mortar to acids’, Civil Engineering journal,
Vol.14, No.8, 38-44. (in Japanese)
V. Pavlik. 1994, ‘Corrosion of cement paste by acetic and nitric acids - Part 1: Calculation of
corrosion depth’, Cement and Concrete Research, Vol.24, No.3, 551-562.
Kurashige, I. 2002, Mechanism and Prediction Method for Deterioration of Concrete Due to Sulfuric
Acid, a doctoral dissertation thesis in Tokyo University, Japan. (in Japanese)
Ueda, H., Takata, J. & Tatematsu, H. 1996, ‘Deterioration mechanism of hardened cement paste by
various acids’, Proceedings of the Japan Concrete Institute, Vol.18, No.1, 879-884. (in Japanese)

TT1-199, Concrete deterioration caused by sulfuric acid attack by K. Kawai, S. Yamaji and T. Shinmi